Block based Method for Real-Time Compound Video Compression

In this paper, we propose a novel compound video compression method for real-time applications of computer screen video transmission. The compound video is a very special kind of video, which usually has less motion and contains a mixture of text, graphics, and natural pictures in each frame. A variety of algorithms has been proposed for compound image compression, which tries to remove the spatial redundancy of a single image. However, few works have addressed the problem of compound videos compression. Therefore, we review the previous works on compound image compression and discuss how to extend the existing algorithms to compound video compression. We propose a new video compression framework based on H.264. In order to improve the visual performance and network efficiency, we propose an adaptive quantization algorithm and a rate control algorithm for compound video compression. The experimental results show that the new compression method is able to compress the compound video in real-time with a relatively low computational complexity and with the motion compensation, the visual quality of the compound video is much better than simple compound image compression.